opsci-toolbox 0.0.11__py3-none-any.whl → 0.0.13__py3-none-any.whl

opsci_toolbox/apis/webscraping.py

@@ -11,6 +11,81 @@ import concurrent.futures
 import pandas as pd
 from tqdm import tqdm
 
+def get_tweet_html(username: str, tweet_id: str, **kwargs) -> str:
+    """
+    Retrieves the HTML code of a tweet given the username and tweet ID.
+
+    Args:
+        username (str): The username of the Twitter account.
+        tweet_id (str): The ID of the tweet.
+        kwargs : additional parameters to pass to the Twitter API.
+
+    Returns:
+        str: The HTML code of the tweet.
+
+
+    """
+    params = {'lang':"en",             # language of the features around the tweet
+              "maxwidth" : 550,        # size of the tweet
+              "hide_media":False,      # to hide photo / video
+              "hide_thread":False,     # to hide original message on replies
+              "omit_script": True,     # to include or not the JS script : <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+              "align": None,           # to align the tweet {left,right,center,none}
+              "theme": "light",        # theme of the tweet {light,dark}
+              "dnt": True              # When set to true, the Tweet and its embedded page on your site are not used for purposes that include personalized suggestions and personalized ads.
+              }
+    
+    params.update(kwargs)
+
+    url = f'https://publish.twitter.com/oembed?url=https://twitter.com/{username}/status/{tweet_id}'
+    response = requests.get(url, params=params)
+
+    if response.status_code == 200:
+        data = response.json()
+        html = data.get('html')
+        return html, username, tweet_id
+    else:
+        print(response.url, "Failed to fetch data from Twitter.")
+        return None, username, tweet_id
+
+
+def parallel_twitter_oembed(usernames, tweet_ids, **kwargs):
+    """
+    Scrapes Twitter oEmbed data for multiple tweets in parallel.
+
+    Args:
+        usernames (list): A list of Twitter usernames.
+        tweet_ids (list): A list of tweet IDs corresponding to the tweets of the given usernames.
+        **kwargs: Additional keyword arguments to be passed to the `get_tweet_html` function.
+
+    Returns:
+        pandas.DataFrame: A DataFrame containing the scraped tweet HTML, username, and message ID.
+
+    Raises:
+        Exception: If there is an error while downloading the tweet HTML.
+
+    """
+    all_data = []
+    with concurrent.futures.ThreadPoolExecutor() as executor:
+        # Submit scraping tasks for each URL and add tqdm progress bar
+        futures = [
+            executor.submit(get_tweet_html, username, tweet_id, **kwargs)
+            for username, tweet_id in zip(usernames, tweet_ids)
+        ]
+        for future in tqdm(
+            concurrent.futures.as_completed(futures),
+            total=len(usernames),
+            desc="Scraping Progress",
+        ):
+            try:
+                data, username, tweet_id = future.result()
+                all_data.append((data, username, tweet_id))
+            except Exception as e:
+                print(f"Error downloading : {e}")
+
+    df = pd.DataFrame(all_data, columns=["tweet_html", "user_name", "message_id"])
+    return df
+
 
 def url_get_domain(url: str) -> str:
     """

opsci_toolbox/helpers/common.py

@@ -310,6 +310,22 @@ def write_pickle(data: pd.DataFrame, path: str, filename: str) -> str:
         pickle.dump(data, f)
     return file_path
 
+def save_df_to_pickle(df: pd.DataFrame, path: str, filename: str) -> str:
+    """
+    Write a DataFrame into a pickle file.
+    
+    Args:
+        data (pd.DataFrame): The DataFrame to be written to the pickle file.
+        path (str): The directory where the pickle file will be saved.
+        filename (str): The name of the pickle file (without the extension).
+    
+    Returns:
+        str: The full path to the saved pickle file.
+    """
+    file_path = os.path.join(path, filename + '.pickle')
+    df.to_pickle(file_path)
+    return file_path
+
 
 def write_list_to_txt(input_list: list, path: str, name: str) -> str:
     """
@@ -1319,13 +1335,13 @@ def categorize_percentiles(percentile: float) -> str:
     Categorizes a percentile value into a string representing its range.
 
     Args:
-    - percentile (float): The percentile value (between 0 and 1).
+        percentile (float): The percentile value (between 0 and 1).
 
     Returns:
-    - str: The category of the percentile value.
+        str: The category of the percentile value.
 
     Raises:
-    - ValueError: If the percentile value is outside the range [0, 1].
+        ValueError: If the percentile value is outside the range [0, 1].
     """
     if not (0 <= percentile <= 1):
         raise ValueError("Percentile must be between 0 and 1 inclusive.")
@@ -1349,4 +1365,160 @@ def categorize_percentiles(percentile: float) -> str:
     elif percentile <= 0.9:
         return '80-90%'
     else:
-        return '90-100%'
+        return '90-100%'
+    
+
+def prepare_data_combinations(df: pd.DataFrame, columns_to_combine : list, col_date : str, date_format : str,  rolling_period : str, col_id : str, col_engagement : str) -> pd.DataFrame:
+    """
+    Prepare data for combinations of columns. Useful for data preparation before dataviz of time series. It adds missing rows for each combination of columns and date. 
+    Args:
+        df (pd.DataFrame): The input DataFrame.
+        columns_to_combine (list): List of column names to combine.
+        col_date (str): Name of the column containing dates.
+        date_format (str): Format of the dates in col_date.
+        rolling_period (str): Rolling period for grouping.
+        col_id (str): Name of the column containing unique IDs.
+        col_engagement (str): Name of the column containing engagement values.
+    Returns:
+        pd.DataFrame: The prepared DataFrame with combinations of columns.
+    """
+
+    df_wt_combinations = df.copy()
+    df_wt_combinations["date"] = pd.to_datetime(df_wt_combinations[col_date], format=date_format).to_numpy()
+
+    # Create all possible combinations of columns indexes
+    # all_combinations = create_combination_index(df_wt_combinations, columns_to_combine, "date", rolling_period)
+
+    # If no columns to combine, just use the date for grouping
+    if not columns_to_combine:
+        df_wt_combinations = (df_wt_combinations
+                              .set_index("date")
+                              .resample(rolling_period)
+                              .agg({col_id: "nunique", col_engagement: "sum"})
+                              .fillna(0)
+                              .reset_index())
+    else:
+        # # Create all possible combinations of columns indexes
+        # all_combinations = create_combination_index(df_wt_combinations, columns_to_combine, "date", rolling_period)
+        
+        df_wt_combinations = (df_wt_combinations
+                              .set_index(["date"])
+                              .groupby(columns_to_combine)
+                              .resample(rolling_period)
+                              .agg({col_id: "nunique", col_engagement: "sum"})
+                              .fillna(0)
+                              .reset_index())
+    
+    return df_wt_combinations
+
+# def create_combination_index(df : pd.DataFrame, columns : list, date_column : str, rolling_period : str) -> pd.MultiIndex:
+#     """
+#     Create all possible combinations of unique values from specified columns and date range.
+    
+#     Args:
+#         df (pd.DataFrame): The input DataFrame
+#         columns (list): List of column names to create combinations from
+#         date_column (str): Name of the date column
+#         rolling_period (str): Frequency for date range (e.g., '1D', '1W', '1M')
+    
+#     Returns:
+#         pd.MultiIndex: MultiIndex with all combinations
+#     """
+#     # Create a list to store unique values for each column
+#     unique_values = []
+    
+#     # Get unique values for each specified column
+#     for col in columns:
+#         unique_values.append(df[col].unique())
+    
+#     # Create date range
+#     date_range = pd.date_range(start=df[date_column].min(),
+#                                end=df[date_column].max(),
+#                                freq=rolling_period)
+    
+#     # Add date range to the list of unique values
+#     unique_values.append(date_range)
+    
+#     # Create MultiIndex from product of all unique values
+#     all_combinations = pd.MultiIndex.from_product(unique_values,
+#                                                   names=columns + [date_column])
+    
+#     return all_combinations
+
+# def prepare_data_combinations(df: pd.DataFrame, columns_to_combine : list, col_date : str, date_format : str,  rolling_period : str, col_id : str, col_engagement : str) -> pd.DataFrame:
+#     """
+#     Prepare data for combinations of columns. Useful for data preparation before dataviz of time series. It adds missing rows for each combination of columns and date. 
+#     Args:
+#         df (pd.DataFrame): The input DataFrame.
+#         columns_to_combine (list): List of column names to combine.
+#         col_date (str): Name of the column containing dates.
+#         date_format (str): Format of the dates in col_date.
+#         rolling_period (str): Rolling period for grouping.
+#         col_id (str): Name of the column containing unique IDs.
+#         col_engagement (str): Name of the column containing engagement values.
+#     Returns:
+#         pd.DataFrame: The prepared DataFrame with combinations of columns.
+#     """
+
+#     df_wt_combinations = df.copy()
+#     df_wt_combinations["date"] = pd.to_datetime(df_wt_combinations[col_date], format=date_format).to_numpy()
+
+#     # Create all possible combinations of columns indexes
+#     all_combinations = create_combination_index(df_wt_combinations, columns_to_combine, "date", rolling_period)
+
+#     # If no columns to combine, just use the date for grouping
+#     if not columns_to_combine:
+#         df_wt_combinations = (df_wt_combinations
+#                               .set_index("date")
+#                               .groupby(pd.Grouper(freq=rolling_period))
+#                               .agg({col_id: "nunique", col_engagement: "sum"})
+#                               .fillna(0)
+#                               .reset_index())
+#     else:
+#         # # Create all possible combinations of columns indexes
+#         # all_combinations = create_combination_index(df_wt_combinations, columns_to_combine, "date", rolling_period)
+        
+#         df_wt_combinations = (df_wt_combinations
+#                               .set_index(["date"])
+#                               .groupby([*columns_to_combine, pd.Grouper(freq=rolling_period)])
+#                               .agg({col_id: "nunique", col_engagement: "sum"})
+#                               .reindex(all_combinations, fill_value=0)
+#                               .reset_index())
+    
+#     return df_wt_combinations
+
+def custom_ordering(df : pd.DataFrame, col_to_order : str, custom_order : list) -> pd.DataFrame:
+    """
+    Orders the values in a DataFrame column based on a custom order.
+    Args:
+        df (DataFrame): The DataFrame containing the column to be ordered.
+        col_to_order (str): The name of the column to be ordered.
+        custom_order (list): The custom order of values.
+    Returns:
+        DataFrame: The DataFrame with the column values ordered according to the custom order.
+    """
+    df[col_to_order] = pd.Categorical(df[col_to_order], categories=custom_order, ordered=True).to_numpy()
+    return df
+
+def calcul_total_et_pourcentage(df : pd.DataFrame, col_gb : list, metrics : dict) -> pd.DataFrame:
+    """
+    Calculates the total and percentage values for the given metrics based on a grouping column.
+    Args:
+        df (DataFrame): The input DataFrame.
+        col_gb (list):  Names of the columns to group by.
+        metrics (dict): A dictionary of metrics to calculate.
+    Returns:
+        DataFrame: The modified DataFrame with total and percentage values added.
+
+    """
+    percentage_agregations = {f'per_{key}': lambda x: x[key] / x[f"total_{key}"] for key in list(metrics.keys())}
+
+    df = (df.join(df.groupby(col_gb)
+                  .agg(metrics)
+                  .add_prefix("total_"), on=col_gb
+                  )
+                .assign(**percentage_agregations).fillna(0)
+        )
+    
+    return df
+

opsci_toolbox/helpers/dataviz.py

@@ -673,19 +673,14 @@ def create_scatter_plot(
     """
     params = general_kwargs()
     params.update(kwargs)
-
     marker_color = params["marker_color"]
     marker_line_color = params["marker_line_color"]
     marker_size = params["marker_size"]
     col_hover = params["col_hover"]
-
     xaxis_range = params["xaxis_range"]
-
     yaxis_range = params["yaxis_range"]
 
-
     fig = go.Figure()
-
     if marker_line_color is None:
         marker_line_color = marker_color
 
@@ -694,40 +689,41 @@ def create_scatter_plot(
         for i, category in enumerate(df[col_category].unique()):
 
             if color_palette:
-                marker_color = color_palette.get(category, generate_random_hexadecimal_color)  # Default to black if category not found
+                marker_color = color_palette.get(category, generate_random_hexadecimal_color())  # Default to black if category not found
             else:
                 marker_color = generate_random_hexadecimal_color()
 
             # hovertemplate generation
-            hovertemplate = (
-                "<b>"
-                + col_x
-                + "</b>:"
-                + df[df[col_category] == category][col_x].astype(str)
-                + "<br><b>"
-                + col_y
-                + "</b>:"
-                + df[df[col_category] == category][col_y].astype(str)
-                + "<br><b>"
-                + col_category
-                + "</b>:"
-                + str(category)
-            )
+            # hovertemplate = (
+            #     "<b>"
+            #     + col_x
+            #     + "</b>:"
+            #     + df[df[col_category] == category][col_x].astype(str)
+            #     + "<br><b>"
+            #     + col_y
+            #     + "</b>:"
+            #     + df[df[col_category] == category][col_y].astype(str)
+            #     + "<br><b>"
+            #     + col_category
+            #     + "</b>:"
+            #     + str(category)
+            # )
+            hovertemplate = ""
             if col_size is None:
                 size = marker_size
             else:
                 size = df[df[col_category] == category][col_size]
-                hovertemplate += "<br><b>" + col_size + "</b>:" + size.astype(str)
+            #     hovertemplate += "<br><b>" + col_size + "</b>:" + size.astype(str)
 
             if len(col_hover) > 0:
                 for c in col_hover:
                     hovertemplate += (
                         "<br><b>"
                         + str(c)
-                        + "</b>:"
+                        + "</b> : "
                         + df[df[col_category] == category][c]
+                        .apply(format_input)
                         .astype(str)
-                        .apply(wrap_text)
                     )
 
             fig.add_trace(
@@ -2038,8 +2034,6 @@ def bar_trend_per_cat(df: pd.DataFrame,
     params = general_kwargs()
     params.update(kwargs)
 
-    col_hover = params["col_hover"]
-
     xaxis_title = params["xaxis_title"]
     yaxis_title = params["yaxis_title"]
     zaxis_title = params["zaxis_title"]
@@ -2060,7 +2054,7 @@ def bar_trend_per_cat(df: pd.DataFrame,
         current_df = df[df[col_cat] == cat]
 
 
-        hovertemplate="<br><b>"+xaxis_title+"</b> :"+current_df[col_x].astype(str)+"<br><b>"+yaxis_title+"</b> - "+current_df[col_y].astype(str)+"<br><b>"+zaxis_title+"</b> : "+current_df[col_z].astype(str)+"<extra></extra>"
+        hovertemplate="<br><b>"+xaxis_title+"</b> :"+current_df[col_x].astype(str)+"<br><b>"+yaxis_title+"</b> - "+current_df[col_y].astype(str)+"<br><b>"+zaxis_title+"</b> : "+current_df[col_z].astype(str)
         # hovertemplate='<b>Categorie : </b>'+str(cat)+'<br><b>Date : </b>'+ current_df[col_x].astype(str) + '<br><b>'+y1_axis_title+'</b> : '+ current_df[col_metric1].astype(str)+' ('+current_df["per_"+col_metric1].map("{:.1%}".format).astype(str)+')' +'<br><b>'+y2_axis_title+'</b> : '+ current_df[col_metric2].astype(int).astype(str)+' ('+current_df["per_"+col_metric2].map("{:.1%}".format).astype(str)+')'
         for c in col_hover:
             hovertemplate += (
@@ -3946,6 +3940,170 @@ def create_radar(df: pd.DataFrame,
     )
     return fig
 
+def bar_subplots_per_cat(df: pd.DataFrame,
+                         col_x: str,
+                         col_y: str,
+                         col_cat: str,
+                         col_stack: str,
+                         color_palette: dict = None,
+                         n_top_words: int = 20,
+                         **kwargs
+                         ) -> go.Figure:
+    """
+    Create subplots of stacked bar charts.
+
+    Args:
+        df (pd.DataFrame): DataFrame containing data for bar charts.
+        col_x (str): Name of the column containing x-axis values.
+        col_y (str): Name of the column containing y-axis values.
+        col_cat (str): Name of the column containing categories.
+        col_stack (str): Name of the column containing stacking values.
+        color_palette (Optional[Dict[str, str]], optional): Dictionary mapping categories to colors. Defaults to None.
+        n_top_words (int, optional): Number of top words to display in each bar chart. Defaults to 20.
+        **kwargs: Additional keyword arguments to update default plotting parameters.
+
+    Returns:
+        go.Figure: Plotly Figure object representing the subplots of stacked bar charts.
+    """
+
+    params = general_kwargs()
+    params.update(kwargs)
+
+    marker_color = params['marker_color']
+    textposition = params["textposition"]
+    vertical_spacing = params['vertical_spacing']
+    horizontal_spacing = params["horizontal_spacing"]
+    col_hover = params['col_hover']
+    n_cols = params['n_cols']
+    categories = df[col_cat].unique()
+
+    # user define a number of columns, we compute the number of rows requires
+    n_rows = math.ceil(len(categories) / n_cols)
+
+    # fine tune parameter according to the text position provided
+    if textposition == 'inside':
+        horizontal_spacing = (horizontal_spacing / n_cols) / 2
+    else:
+        horizontal_spacing = (horizontal_spacing / n_cols)
+
+    # create subplots
+    fig = make_subplots(
+        rows=n_rows,  # number of rows
+        cols=n_cols,  # number of columns
+        subplot_titles=list(categories),  # title for each subplot
+        vertical_spacing=vertical_spacing / n_rows,  # space between subplots
+        horizontal_spacing=horizontal_spacing,  # space between subplots
+        shared_xaxes=params["shared_xaxes"],
+        shared_yaxes=params["shared_yaxes"]
+    )
+
+    # create stacked bar traces for each subplot
+    row_id = 0
+    col_id = 0
+    for i, category in enumerate(categories):
+        # define row and column position
+        col_id += 1
+        if i % n_cols == 0:
+            row_id += 1
+        if col_id > n_cols:
+            col_id = 1
+
+        # select data
+        current_df = df[df[col_cat] == category].sort_values(by=col_x, ascending=True)
+        unique_stacks = current_df[col_stack].unique()
+
+        if textposition == 'inside':
+            text = current_df[col_y].head(n_top_words)
+        else:
+            textposition = "auto"
+            text = None
+
+        for stack in unique_stacks:
+            # define bar color or create a random color
+            if color_palette:
+                marker_color = color_palette.get(stack, generate_random_hexadecimal_color())
+            else:
+                marker_color = generate_random_hexadecimal_color()
+                
+            stack_df = current_df[current_df[col_stack] == stack]
+            hovertemplate = '<b>'+col_cat+" : "+ stack_df[col_cat].astype(str)+ '</b><br>' + col_stack+" : "+ stack_df[col_stack].astype(str)
+            
+            for col in col_hover:
+                hovertemplate += '<br><b>' + col + ': ' + current_df[current_df[col_cat] == category][col].astype(str) + '</b>'
+
+
+            fig.add_trace(
+                go.Bar(
+                    x=stack_df[col_x].tail(n_top_words),
+                    y=stack_df[col_y].tail(n_top_words),
+                    opacity=params["marker_opacity"],
+                    orientation=params["orientation"],  # horizontal bars
+                    name=stack,  # trace name for legend
+                    text=text,  # text to display
+                    textposition=textposition,  # text position
+                    textangle=params["xaxis_tickangle"],  # text angle
+                    marker_color=marker_color,  # bar color
+                    hovertemplate=hovertemplate + "<extra></extra>"  # hover info
+                ),
+                row=row_id,
+                col=col_id
+            )
+
+    for row_id in range(1, n_rows+1):
+        for col_id in range(1, n_cols+1):
+            fig.update_yaxes(title=params["yaxis_title"], row=row_id, col=1)
+            fig.update_xaxes(title=params["xaxis_title"], row=row_id, col=col_id)
+
+    fig.update_layout(
+        margin=dict(l=75, r=75, t=75, b=50),
+        title_text=params["title_text"],
+        width=n_cols * params["width"],  # plot size
+        height=n_rows * n_top_words * params["height"],  # plot size
+        showlegend=params["showlegend"],
+        font_family=params["font_family"],
+        font_size=params["font_size"],
+        template=params["template"],
+        plot_bgcolor=params["plot_bgcolor"],  # background color (plot)
+        paper_bgcolor=params["paper_bgcolor"],  # background color (around plot)
+        uniformtext_minsize=params["uniformtext_minsize"],
+        barmode=params['barmode']
+    )
+
+    fig.update_yaxes(
+        # title=params["yaxis_title"],
+        title_font_size=params["yaxis_title_font_size"],
+        tickangle=params["yaxis_tickangle"],
+        tickfont_size=params["yaxis_tickfont_size"],
+        range=params["yaxis_range"],
+        showgrid=params["yaxis_showgrid"],
+        showline=params["yaxis_showline"],
+        zeroline=params["yaxis_zeroline"],
+        gridwidth=params["yaxis_gridwidth"],
+        gridcolor=params["yaxis_gridcolor"],
+        linewidth=params["yaxis_linewidth"],
+        linecolor=params["yaxis_linecolor"],
+        mirror=params["yaxis_mirror"],
+        layer="below traces",
+    )
+
+    fig.update_xaxes(
+        # title=params["xaxis_title"],
+        title_font_size=params["xaxis_title_font_size"],
+        tickangle=params["xaxis_tickangle"],
+        tickfont_size=params["xaxis_tickfont_size"],
+        range=params["xaxis_range"],
+        showgrid=params["xaxis_showgrid"],
+        showline=params["xaxis_showline"],
+        zeroline=params["xaxis_zeroline"],
+        gridwidth=params["xaxis_gridwidth"],
+        gridcolor=params["xaxis_gridcolor"],
+        linewidth=params["xaxis_linewidth"],
+        linecolor=params["xaxis_linecolor"],
+        mirror=params["xaxis_mirror"],
+        layer="below traces"
+    )
+    return fig
+
 # def bar_subplots(df: pd.DataFrame,
 #                  col_x: str,
 #                  col_y: str,

opsci_toolbox/helpers/dates.py

@@ -77,3 +77,49 @@ def df_col_to_datetime(df: pd.DataFrame, col: str) -> pd.DataFrame:
     df[col] = pd.to_datetime(df[col])
     return df
 
+
+# from dateutil import parser
+# from datetime import datetime
+
+# def detect_date_format(date_string):
+#     formats = [
+#         # Date formats
+#         "%Y-%m-%d", "%d-%m-%Y", "%m/%d/%Y", "%m-%d-%Y", 
+#         "%Y/%m/%d", "%d/%m/%Y", "%Y.%m.%d", "%d.%m.%Y",
+#         "%d %b %Y", "%d %B %Y", "%b %d, %Y", "%B %d, %Y",
+#         "%d-%b-%Y", "%d-%B-%Y", "%b-%d-%Y", "%B-%d-%Y",
+#         # Date and time formats
+#         "%Y-%m-%d %H:%M:%S", "%d-%m-%Y %H:%M:%S", "%m/%d/%Y %H:%M:%S", "%m-%d-%Y %H:%M:%S",
+#         "%Y/%m/%d %H:%M:%S", "%d/%m/%Y %H:%M:%S", "%Y.%m.%d %H:%M:%S", "%d.%m.%Y %H:%M:%S",
+#         "%d %b %Y %H:%M:%S", "%d %B %Y %H:%M:%S", "%b %d, %Y %H:%M:%S", "%B %d, %Y %H:%M:%S",
+#         "%d-%b-%Y %H:%M:%S", "%d-%B-%Y %H:%M:%S", "%b-%d-%Y %H:%M:%S", "%B-%d-%Y %H:%M:%S",
+#         # Time formats with milliseconds
+#         "%Y-%m-%d %H:%M:%S.%f", "%d-%m-%Y %H:%M:%S.%f", "%m/%d/%Y %H:%M:%S.%f", "%m-%d-%Y %H:%M:%S.%f",
+#         "%Y/%m/%d %H:%M:%S.%f", "%d/%m/%Y %H:%M:%S.%f", "%Y.%m.%d %H:%M:%S.%f", "%d.%m.%Y %H:%M:%S.%f",
+#         "%d %b %Y %H:%M:%S.%f", "%d %B %Y %H:%M:%S.%f", "%b %d, %Y %H:%M:%S.%f", "%B %d, %Y %H:%M:%S.%f",
+#         "%d-%b-%Y %H:%M:%S.%f", "%d-%B-%Y %H:%M:%S.%f", "%b-%d-%Y %H:%M:%S.%f", "%B-%d-%Y %H:%M:%S.%f",
+#         # ISO format
+#         "%Y-%m-%dT%H:%M:%S", "%Y-%m-%dT%H:%M:%S.%f",
+#         "%Y-%m-%d %H:%M:%S", "%Y-%m-%d %H:%M:%S.%f",
+#         "%Y-%m-%dT%H:%M:%S%z", "%Y-%m-%d %H:%M:%S%z",
+#         # Additional formats
+#         "%y/%m/%d %H:%M:%S", "%d/%m/%y %H:%M:%S", "%y-%m-%d %H:%M:%S", "%d-%m-%y %H:%M:%S",
+#     ]
+
+#     for date_format in formats:
+#         try:
+#             # Try to parse the date string with each format
+#             parsed_date = datetime.strptime(date_string, date_format)
+#             return date_format
+#         except ValueError:
+#             continue
+
+#     return None
+
+# def detect_date_format(date_string):
+#     try:
+#         # Use dateutil parser to parse the date string
+#         parsed_date = parser.parse(date_string, fuzzy=False)
+#         return parsed_date
+#     except ValueError:
+#         return None